We had halted on a lovely island—Adelaide Island—with a rocky and sandy extension. The night of August 5th had been stifling, with a minimum temperature of 72° F.

I found my work too much for me now. There was too much to observe on all sides. We were travelling quickly with the swift current. A hill range from east to west, 300 ft. high, ran along the left bank. Farther, where the river went to the north-east for 4,000 m., laminated rock like slate showed through the left bank, especially in a semicircular indentation which had been eroded by the water. There a strong whirlpool had formed. Another great stretch of river, 5,500 m., was now before us, with a small hill 80 ft. high on the right bank. The river next formed a circular basin with three islets and a barrier 500 m. across.

We were now in a region where, fortunately for us, castanheiro trees (vulgo. the "Para chestnut") were to be found. Fish was scarce in the river. Now that we had almost superhuman work to accomplish, our meals were extremely scanty owing to the loss of our provisions, and we had not sufficient food to keep up our strength.

As we went on I saw to the north-east of us another hill-range 300 ft. high, extending from north-west to south-east, like most of the ranges found in that region. Where a prominent headland stood on the left side, with a hill 250 ft. high upon it, the river turned to 30° b.m. The hill was made up of foliated rock lying in strata that varied from one inch to one foot in thickness.

On the right side of the stream great cubic blocks of rock rested on the polished curves of a huge dome of granite. A quantity of dÉbris stretched from south to north right across the basin, and caused a deviation in the stream.

Conveying the Canoe across the Forest on an Improvised Railway and Rollers.

Conveying the Canoe across the Forest on an Improvised Railway and Rollers.

A terrific rapid with a sheer drop of 3 ft. was situated here. A double whirlpool of great magnitude was formed at the bottom of the rapid, the water revolving with such force that the concavity was gradually depressed for some 3 ft. and had a great hole in each centre. We shot that rapid. As Alcides on that occasion followed my instructions, the canoe shot past between the two whirlpools, and although even then she nearly capsized, we were able to continue, my men shrieking with merriment at what they now believed to be their invulnerability. We dodged the unpleasant eddies while we floated with great speed in the strong current.

The river, which had contracted that day to 250 m., now expanded once more into a large basin 1,200 m. wide and 1,800 m. long, with most troublesome eddies as we went through it. The river described a great turn from N.N.E. to 180° b.m. or due south. To add to the pleasures of our existence, we came in for a heavy rain-storm that day, with deafening thunder and blinding lightning. Notwithstanding the great discomfort it caused us, it pleased me very much because of the wonderful effects of light it produced on the river.

Where the stream, in a course which had wriggled like a snake, turned once more due north to 360° b.m., it divided itself into two small channels. High waves were produced where the water, pushed by the wind, was forced against the rapid. There was a good drop in the level of the river at that rapid, and it was a nasty place indeed for us to go through. We got tossed about, splashed all over, but we came out of it all the same, amid the wildly excited yells of my men. They were beginning to think that they were the greatest navigators that had ever lived, and they never let an opportunity pass of reminding each other of that fact.

I halted in the middle of the day to take the usual observations for latitude and longitude (lat. 8° 47'·5 S.; long. 58° 39' W.), but I was interrupted in my work by another heavy rain-storm, which came and drenched us once more. After that dense clouds as black as ink covered the entire sky for the whole afternoon. We were now in the rainy season. Terrific gusts preceded these rain-storms, and were most troublesome to us.

After negotiating the bad rapids, the river went through a basin of boulders of broken foliated rock. There were three small channels. Then beyond, the entire river was forced through a rocky channel from 35 to 40 m. wide, the water rushing through with incredible force on a steep gradient until half-way down the channel, where it actually ran uphill for 50 m. or so, so great was the impetus it had received on its rapid descent to that point.

You can well imagine what a pleasant job it was for us to convey the canoe along with ropes over so delightful a spot. Owing to our insufficient food, our strength had greatly diminished. The ropes we had used on the many rapids were now half-rotted and tied up in innumerable knots. Moreover, the banks of sharp cutting rock were of great height, and our ropes were not long enough to be used separately, so that we decided to use only one long rope made up of all the ropes we possessed tied together. To make matters more difficult, the channel was not perfectly straight, but described two or three sharp corners, where the water was thrown with much vigour in one direction, then, being driven off immediately at a different angle, curled over itself, producing mountains of foaming water forty or fifty feet in height, and leaving great depressions near the inner corner.

We cut down some long poles, and I placed one man with a big pole on guard at each corner close to the water, in order to push the canoe away toward the middle of the stream in case she came too near those dangerous points.

That channel was some 600 m. long. When we were ready we let the canoe go, all spare hands holding fast to the rope, running and scrambling up and down and along the high rocky cliff, the canoe giving us violent jerks when the direction of the current was changed. With much alarm we saw her spring up in the air like a flying-fish on one or two occasions. We ran along like mad, out of breath and sweating, trying to keep ahead of the canoe. The two men with poles also ran along after the danger points were passed, so as to shove her along when she came too near other dangerous rocks.

After a race of great excitement, we all, with bleeding feet and hands—the palms of our hands actually blistered by the rope which slid through our tightly closed fists—were eventually able to pull the canoe safely on shore below the rapid.

In that mad flight I found time to pull out the camera for one second and take a snapshot of the canoe in the middle of the rapid. The photograph is reproduced among the illustrations of this volume.

My men were so tired that it was impossible to go on. Moreover we had before us the second section of that formidable rapid, and we could not negotiate this without emptying the canoe, which was full of water, and readjusting the rope.

We spent the night of August 6th on those rocks, the minimum temperature being 63° F.

When we went on with our dangerous work the next morning we had the greatest difficulty in saving the canoe, as in entering the whirlpool she was swamped, and it was all we could do to pull her back towards the bank before she foundered altogether. The actual drop in that rapid was not less than 8 ft. vertically. We just managed to rest her on a submerged rock until we were able to bale some of the water out.

That canoe was really wonderful in a way. My men patted her on the prow as if she had been an animal, and said she was a good canoe. Indeed she was, but in her old age she felt the strain of that exciting journey. Every time I looked at her I did not know how much longer she might last. Whatever may be said of them, my men must be given credit for their courage in going along in that canoe. I do not believe that there are six other men in Brazil—or perhaps in any other country—who would have ventured to go across even the most placid pond in a similar craft.

After the rapids came a great basin 1,000 m. long, 800 m. wide. There the river described an angle from 20° b.m. to 45° b.m., and we perceived two parallel ranges before us to the N.N.E., the farther one much higher than the one nearer. Some 5 kil. beyond was yet another rapid, but not so troublesome a one this time. The river there diverged from north-east to a direction due west. A hill range, from 150 to 250 ft. high, extended from W.S.W. to E.N.E. An isolated hill, 300 ft. high, could be seen to the E.N.E.

We suffered agony that day from regular clouds of borrachudos, terrible little sand mosquitoes which made life an absolute burden in that region. Our faces, arms, and legs were a mass of ink-black marks left by the stings of those vicious brutes. Particularly when our hands were occupied in holding the canoe going down rapids, or busy with dangerous jobs, did swarms of those little rascals attack us with indomitable fury.

Pushing the Canoe Uphill through the Forest.

Pushing the Canoe Uphill through the Forest.

(Notice men with heads wrapped owing to torturing insects.)

Another basin was met, 700 m. wide, quite shallow, and with rapids over a barrier of rock extending across it from south-west to north-east. That barrier was most interesting, because in many places great lava-flows were visible; in other places masses of ferruginous rock could be observed, with most extraordinary patterns upon them—triangles, rectangles, trapeziums, and all kinds of other angular geometrical patterns, such as we had met before on the high plateau of Matto Grosso.

We stopped in the middle of the day on an island 1,200 m. long, from which we obtained a fine view of the hill range looming before us from W.S.W. to E.N.E. on the right bank.

I was having great trouble with my chronometer, which the many jerks, falls, and baths did not seem to improve. I checked it whenever I could by observations of local time and by other watches which I carried. But all my instruments were beginning to feel the effects of that journey very much. The wonder to me was that they had got so far in as good condition as they were, considering all we had gone through.

Our lunch was speedy, as we had nothing to eat. The moment I had finished my observations for latitude and longitude we started off once more, my men keeping their eyes all the time on the forest on the look-out for nut-trees, the river that day giving us no fish at all.

Within ten minutes we had shot two powerful rapids, and in one place went over a dangerous submerged wall of rock extending across the river from E.S.E. to W.N.W.

The men—very hungry—were extremely quarrelsome that day and insulting to one another. The canoe went broadside down a rapid we met, the men gesticulating instead of paddling along as they should have done. With a great bump we stuck with a heavy list to starboard on a rock in the middle of the rapid, and presently the canoe was filled with water. Had we not stuck fast on that rock we certainly should have capsized. The water was baled out in due course, the canoe was floated once more. Soon afterwards another strong rapid, with a pedraria extending right across the stream from S.S.W. to N.N.E., gave us endless trouble.

I warned Alcides to get us alongside some rocks in order that we might let the canoe down with ropes, as the rapid, with a sheer drop of over 6 ft., looked too dangerous for us to shoot it. But Alcides was furious with the other men, and in order to punish them steered the canoe into the most dangerous part of the rapid. A second later the canoe, at an angle of 45°, was swept away down the foaming current along the slant of the rapid, which extended there for about 15 m. The channel was a most intricate one, with rocks scattered all over it, so that it was absolutely impossible for the canoe, with her great length, to go through without having an accident.

As we shaved a big rock in the middle of the rapid, and I saw the canoe steering straight for another big rock in front, I knew disaster was imminent, and leapt out on the rock. So frequently was it necessary for me to do so, that I had become quite an expert at jumping, and had acquired almost the agility of a monkey. Alcides, too, seeing the danger, also tried to follow my example, but unfortunately missed his footing and was swept away by the current. I just managed to seize him before he disappeared for good, and dragged him safely on to the rock.

In the meantime the canoe had swung with great vigour and struck the big rock sideways, smashing her side and filling at once with water. All the baggage was swamped; only a portion of the canoe aft remained above the water, many of our things being washed away altogether.

There she stuck, fortunately for us. With considerable danger we managed to undo the ropes which were fastened to her stern. After several hours of hard work—and of extreme peril for the men who could not swim, as we had to work all the time with the water up to our necks in a powerful current, which made it most difficult to keep our footing—we succeeded in pulling her off and taking her alongside the bank.

That disaster was rather a serious one for us, as it injured many of my instruments, particularly the aneroids; but I considered myself fortunate in managing to save all the photographs and notebooks as well as the instruments for taking astronomical observations, which were kept in airtight cases. I lost my favourite pair of shoes, which were by my side in the canoe when I jumped out.

As it so frequently happened that we had to jump into the water—in fact, we spent more time in the water than out—I had adopted as a costume my pyjamas, under which I always wore the belt with the heavy packages of money. The paper money—a very considerable sum—had with the many baths become a solid mass. I could not well spread the banknotes out in the sun to dry, as I did not wish my men to know how much I possessed; so that for many, many weeks I had around my waist those heavy leather wallets soaked in water, my natural heat not being quite sufficient to dry them.

We had worked in the stream until nearly midnight. We had nothing to eat when we had finished our work, and the result was that the next morning my men were still tired.

Two of my cameras were by my side when the canoe was swamped, one containing eighteen plates, the other twelve, all of which had been exposed. The cameras, being heavy, remained at the bottom of the canoe and were saved, but the bath did not do them good. I did not want to lose the plates, so there was only one course to follow, and that was to develop them while they were still wet. While my men slept I sat up a good portion of the night developing all those plates—quite successfully too—and trying to clean and fix up the cameras again for use the next day. One of my other cameras had been destroyed previously by one of my men, who sat on it, and of course smashed it to pieces. Another camera, which was still in excellent condition, having been in an air-tight case, was rather too big to be used for the work in going down the rapids.

During the night of August 7th the minimum temperature was 62° F.

Conveying the Canoe, weighing 2,000 lb., over a Hill Range—The Descent.

Conveying the Canoe, weighing 2,000 lb., over a Hill Range—The Descent.

I worked the entire morning with Alcides, trying to mend the poor canoe. The hole which had been made in her side was so big that Alcides could insert his head into it with great ease. It was not until two o'clock in the afternoon that we started once more. Along the river, which flowed in that particular section to the south-west, was a hill range on the north-west. The range rose 300 ft. above the level of the river. We had gone only some 2,000 m. when we came to another bad rapid stretching across the river from south-east to north-west. We were in a hilly region, hills being visible all along the stream. Soon afterwards we came to another powerful fall over a vertical rocky wall extending from north-west to south-east. Such redoubtable waves were produced there by the force of the water shooting over and then rebounding upwards, that we had to use the greatest care in letting down the unloaded canoe. At one moment she was more than two-thirds out of the water, only her stern resting on the top of the fall, the rest projecting outward in the air for some moments until she dropped down again.

Since the day we had taken the canoe over the hill range at the August Falls, I had doubled my men's salaries—although their original salaries were already many times higher than they would receive from Brazilian employers. I fully recognized that the work was hard, and I wished to encourage them in every possible way.

Next, the river went through a narrow gorge, only 80 m. wide, where the current was mighty strong. High volcanic rocks stood on the right side of us. When we emerged from the narrow neck, which measured some 500 m. in length, we found powerful whirlpools. Farther on the river once more went through a bad narrow passage, 40 to 60 m. wide, with a succession of rapids—extremely unpleasant—for a length of 600 m.

My men were in great form that day, and we shot one rapid after another in fine style, Alcides—for a change—being amenable to reason and following my instructions, which carried us through that dangerous section without mishap. The stream Uruguatos entered the Arinos just above the latter rapids.

That day was indeed a trying one for us. Another narrow channel, 50 m. wide, was reached, along the 250 m. length of which we proceeded with great caution. Then a big basin spread out before us, where the current and eddies were terrific. The bottom of the river was mostly rocky, with great holes and depressions which caused the water to rotate in all directions. In some places amidst the foaming waters could be seen great circles of leaden-looking water, as still as oil. It was in a similar place in the Niagara whirlpool that the famous swimmer, Captain Webb, disappeared for ever. We saw thousands of those places on the Arinos.

The line of the banks on both sides was extremely rocky. In front of us we had a hill with extensive campos on its northerly slope. Then we came to the next rapid. We had endless trouble in this rapid, followed by a second one, practically a continuation of the first.

For 1,000 m. the navigation was extremely dangerous. We unloaded and reloaded the canoe dozens of times that day, although the work of taking the baggage over on our heads was not so troublesome now, as we had very little baggage left. But if we had not much, it was still the heaviest cases which remained. All together they weighed between five and six hundred pounds. The river ran beside a range of hills on the left side.

When we halted, exhausted, late at night we had travelled that day the meagre distance of 9,900 m.

My men killed two large spider monkeys, which supplied them with a meal. I could not touch them, as the monkeys looked too human for words. It made me positively ill to see one of my men biting with great gusto at an arm and hand which had been roasted on the flames, and which looked exactly like a portion of a human corpse. The smell, too, of the roasted monkeys was similar to the odour of roasted human beings—which I knew well, as I had on several previous occasions been at rough cremations of people in Japan, in the Himahlya (or Himalayas), and in Africa.